Wiring and powering GPUs differently can swing AI latency by orders of magnitude, says CoreWeave

What happened

Wiring and powering GPUs (a chip built for many calculations at once, used for graphics and AI) differently can swing AI latency (the delay between asking for something and getting it) by orders of magnitude, says CoreWeave, according to SiliconANGLE. The neocloud market is moving past its origins as a stopgap for scarce graphics processing units. AI-native startups now choose their infrastructure on latency, burst capacity and openness, not just chip availability.

That shift is playing out at CoreWeave Inc., which is expanding beyond GPU compute into networking, storage and software as inference demand grows. UPDATED 07:49 EDT / OCTOBER 02 2026 INFRA Wiring and powering GPUs differently can swing AI latency by orders of magnitude, says CoreWeave by Ryan Stevens The neocloud market is moving past its origins as a stopgap for scarce graphics processing units.

has evolved from an open-source framework for retrieval-augmented generation into a model builder that rents its compute rather than owning it, according to Jerry Liu (pictured, right), co-founder and chief executive officer of LlamaIndex. Liu and Lukas Biewald (left), senior vice president of AI initiatives at CoreWeave, spoke with theCUBE’s John Furrier and Dave Vellante at Fully Connected , during an exclusive broadcast on theCUBE, SiliconANGLE Media’s livestreaming studio.